Sub-Saharan Africa has over 80% of the world's HIV infections, and an estimated 70% of prevalent and incident infections in African cities are the result of transmission between married partners. Our research team has 15 years' experience conducting heterosexual HIV prevention research in the capitals of two African countries: Kigali, Rwanda (1986-) and Lusaka, Zambia (1994-). Our early studies in Rwanda showed that Voluntary HIV counseling and testing (VCT) results in pronounced risk reduction in married couples. Since we first published these findings in 1991, we have provided VCT to over 20,000 Rwandan and Zambian couples. In the last year we have implemented 'same-day' couples' VCT in antenatal care clinics in order to combine prevention of mother-to-child transmission and heterosexual HIV transmission. Couples' VCT is the only prevention program proven to reduce HIV incidence in the largest risk group in the world, African couples. It is also feasible, cost-effective, and popular with clients. In spite of this, in 2002, twenty years after the start of the HIV epidemic, ten years after the first publication about couples' VCT in Africa, and three years after the discovery of inexpensive regimens for perinatal HIV prevention, VCT has been provided to < 1% of African couples. A number of structural factors affect demand for and supply of couples' VCT. Demand for couples' VCT is low, because of the belief that monogamy is 'safe', the fear of stigma, gender inequity between husband and wife, and lack of knowledge about where VCT can be obtained. Given the low demand, policymakers and other influential groups have not promoted couples' VCT. In turn, funding agencies have not supported VCT services, further compromising supply and ensuring low utilization. Given what we know about the beneficial impact of couples' VCT, it is critical that this continuing cycle of low demand-low supply be interrupted. We propose three linked activities in each of two capital cities in Africa: a) a quasi-experimental three armed study of community based interventions to increase couples' Voluntary HIV testing and counseling (specific aims 3 and 4); b) follow-up behavioral assessments of structural factors in HIV discordant couples, with the goal of improving couple counseling strategies (specific aim 5); and c) systematic inclusion of community influence networks, policymakers, and funding agencies in the development and implementation of the interventions, a process that will culminate in a plan and commitment for sustainability and dissemination of couples' VCT (specific aims 1, 2, and 6).